Aspects of the germination of some New Zealand

Simon H. Moore1 and Peter Bannister

Department of Botany, University of Otago, P.O. Box 56, Dunedin, New Zealand 1Current address: Department of Conservation, Nelson/Marlborough Conservancy, Private Bag 5, Nelson, New Zealand

Abstract Seeds of Northern Hemisphere ericoid are light-sensitive, long-lived, form seed banks, and may be stimulated by short periods of high temperature, but are usually unaffected or show reduced germination in response to low temperature stratification. Comparative study of the germination responses of five New Zealand species of Gaultheria (G. antipoda, G. crassa, G. depressa, G. macrostigma, G. parvula), two hybrids (G. macrostima x crassa, G. macrostigma x depressa) and one introduced European species ( lusitanica) showed that light stimulated germination of all species, while germination was not stimulated in dry seed subjected to short periods of high temperature but viability was little affected. Germination of imbibed seed of two other European species (Erica ciliaris, E. erigena) was stimulated by short exposures to high temperature, and germination in E. erigena was inhibited by low temperature stratification. Germination of some New Zealand seed lots was stimulated by low temperature stratification, particularly in collections from high altitude (G. depressa) and in some collections of G. macrostigma and its hybrid with G. depressa. Seeds were largely unaffected by up to ten successive periods of hydration and desiccation, and remained viable after 16 months of dry storage and 12 months of moist storage. There was evidence of after-ripening in seed of G. depressa and G. macrostigma x depressa. Seed lots collected from the same sites in different years showed variable germination responses. Seed germination in Ne"w Zealand Ericaceae is perhaps more similar to Northern Hemisphere Ericaceae than it is to other native woody .

Additional key words: Erica, Gaultheria, Pemettya, after-ripening, desiccation, heat shock, light/dark, low temperature, stratification

low temperature stratification and germination may be Introduction stimulated by short periods of-i exposure to high Ericaceous plants have been widely studied in temperatures (Whittaker and Gimingham, 1962; Ban­ northern and western Europe, largely because of interest nister, 1965; Mallik and GiminghaDi, 1985) and diurnal in heathland, initially as a wasteland that could be made fluctuations in temperature (Gimingham, 1960). more productive and more recently as a disappearing In New Zealand, native ericaceous species are habitat that needs to be preserved (Gimingham, 1992). confined to species of Gaultheria and Pernettya. These Consequently the biology and ecology of British and two genera are differentiated on the basis of fruit European heath species has been well studied and the characters: in Gaultheria a dry capsule is surrounded by germination characteristics of European ericaceous heath a fleshy calyx, whereas in Pernettya the true fruit is a plants are reasonably well known, particularly for ericoid berry with a persistent calyx which may also be species such as heather (Calluna vulgaris) and common succulent. Seeds are numerous and small, and of a species of Erica (E. cinerea, E. tetralix). similar size to those of European ericoid species (Moore, These species have small (< 1 mm in length), long­ 1996). Intergeneric hybrids producing fertile seed occur lived seed that forms persistent seed banks. Afforest­ and, consequently, we have followed Middleton (1991) ation of heathland in Britain leads to the elimination of in this paper and included all Pernettya species within heathland species, such as Calluna vulgaris and Erica Gaultheria. Written information on the germination tetralix, but there is significant regeneration of these requirements of Gaultheria species native to New species when trees are felled after 40 years or more (Hill Zealand appears to be limited to the unpublished work of and Stevens, 1981). As with many species that form Armstrong (1964), a short paper by Bannister (1990) and seed banks, seeds do not germinate in the dark but are brief mentions elsewhere (e.g., Fountain and Outred, stimulated by exposure to light, do not appear to require 1991; Bannister and Jarneson, 1994). Gaultheria species

Seed Symposium 1999 83 Germination aspects of some NZ Ericaceae show no or minimal germination in the dark but germin­ seeds were stored either in light-proof canisters (E. ation is stimulated by exposure to light and (in G. lustitanica) or in transparent containers (E. ciliaris, E. macrostigma) may be enhanced by low temperature strat­ erigena) at room temperature. ification (Bannister, 1990; Bannister and Jameson, 1994). Germination of the South American G. mucronata Seed germination appears to respond to low temperature stratification Seeds were plated out on two layers of Whatman No (Arena et al., 1994). The germination ecology of Erica 29 filter paper placed over a layer of absorbent cotton lusitanica, a European ericoid species naturalised in New wool in glass petri dishes. Paper and cotton wool were Zealand, has been investigated by Mather and Williams kept moist with distilled water during the course of the (1990). This paper also presents data for seeds of two experiments. Only apparently filled seeds were used in other European species, Erica erigena and E. ciliaris, the experiments and typically two replicates of 100 seeds collected in the British Isles and germinated in New were used for each treatment. In experiments with E. Zealand. The data for E. ciliaris have been published erigena, only 50 seeds were used in each replicate and elsewhere (Rose et al., 1996). only 25 seeds were used in hydration-dehydration The object of this work was to examine the experiments. germination responses of native Gaultheria species in Seeds of the Gaultheria species and Erica lusitanica more detail and compare them with what is known for were kept in an illuminated cabinet (50 J.tmollm2/s) with Northern Hemisphere Ericaceae. The principal factors a twelve hour alternation of light at 22°C and dark at examined in this paper are responses to high and low l2°C. Petri dishes were laid out in a randomised block temperature, desiccation, after-ripening and longevity. design within the cabinets. Some treatments were started in, or transferred to, an unheated, shaded (c. 56 % of ambient) glasshouse. Seeds of Erica species were ger­ Materials and Methods minated at ambient laboratory temperatures and light. Species and sites of seed collection Radicle emergence was taken as evidence of germination Seed of Erica lusitanica was collected from the and experiments were terminated after about 270 d when Silverpeaks, near Dunedin, in April1992 and April1993, germination had usually ceased, although some experi­ that of E. ciliaris from Dorset, England, in October 1994 ments with Erica erigena were terminated after one year. and that of E. erigena from Errisbeg, Co. Galway, and The final germination percentage was the main statistical Mallaranny and Furnace Lough, Co. Mayo, Ireland, in parameter used in analyses of variance. The inverse of August 1994. Seeds of native New Zealand species were the time taken for 50 % of final germination was also collected from a variety of sites in the southern South used as a function of germination rate, but is referred to Island of New Zealand (site details are given by Moore in the text only if there is a particular point to be made. 1996). In April 1992 and 1993 collections of seed were made from Leith Valley, Dunedin (Gaultheria antipoda) Experimental treatments and the Hawkdun Range, near Undaunted Creek and on Seeds were stratified at 4°C for four weeks, although Otematata Station (G. crassa, G. depressa from 775 m, a 12 week period was also used for Gaultheria 1280 m, 1580 m, G. macrostigma (=Pernettya macro­ macrostigma and G. macrostigma X depressa, and stigma) and G. parvula (=Pernettya nana)). Seeds of G. additional periods of 8 and 12 weeks for E. ciliaris and macrostigma and itshybrids were collected from Flag­ E. erigena. Heat treatments of 30 sec and 2 min at staff, Dunedin, in April 1992 and March 1993 (G. 100°C were used on dry seeds of the Gaultheria species macrostigma, G. macrostigma X depressa, G. macro­ and E. lusitanica, whereas imbibed seeds of the Erica stigma x crassa) with additional collections of G. macro­ ciliaris and E. erigena were soaked in water at 20, 50, stigma in December 1992 and May 1993. and 80°C for five min and at 100°C for one and five Mature fruits were collected from several plants min. within each site. The fleshy fruits of the Gaultheria After-ripening and longevity of seed was tested by species were placed in petri-dishes and allowed to dry in germinating seed stored dry for 1, 3 and 16 months. The the sun. Dried fruits were broken open to release the response of seeds to repeated desiccation and re­ seeds which were separated from any extraneous material hydration was tested for up to ten cycles where seeds and stored in light-proof canisters at room temperature were allowed to imbibe for a week and then desiccated until used. The fruits of the Erica species were already for 24 h before being rehydrated. air-dry on collection and further drying caused their The effects of time of harvest on seed of G. capsules to open and shed their seed. The separated macrostigma collected from Flagstaff at various times

Seed Symposium 1999 84 Germination aspects of some NZ Ericaceae during summer and autumn (1992-1993) were also continuous germination through the experimental period. evaluated by both standard germination in the cabinet Further details of germination patterns are given in and by using seed that had been stratified at 4°C for 12 Moore (1996). weeks. Ecotypic comparisons were also made on G. macrostigma from various sites and G. depressa from an Germination in the growth cabinet and in the altitudinal range. Comparisons were also made with glasshouse hybrid seeds and their parents. In all species (including Erica spp.), germination was greatest in the light and, except for two collections of G. macrostigma where there was a minimal amount of Results germination, there was no germination in the dark Patterns of gennination (Moore, 1996). E. lusitanica showed low germination in Erica lusitanica showed only low germination in the the low illumination of the cabinet but germination was cabinet (< 20 %) which was completed in about fifty substantially enhanced in the glasshouse, particularly in days. However, after six months to a year, when petri­ seed that had had a pretreatment of 375 days of moist dishes from various completed experiments were storage in the dark. Germination of Gaultheria antipoda transferred from the cabinet to the glasshouse, a and G. depressa was more or less unaffected by secondary burst of germination occurred resulting in 70- conditions; G. crassa showed poorer germination after 90 % final germination (Moore 1996). Seed of Erica the dark pretreatment, and G. macrostigma and G. ciliaris and E. erigena was germinated in laboratory parvula showed reduced germination in the cabinet conditions with uncontrolled fluctuations in light and (Table 1). temperature andphotoperiod. Germination was low in E. ciliaris (20-30 o/o) and more or less complete after 50 Germination in response to temperature days. In Erica erigena germination commenced only pretreatments after one month with a major spurt of germination In general, the species examined showed little or no around 100 days (autumn) and a secondary spurt around response to low temperature stratification. In Erica 200 days (midwinter). Maximum germination in untreat­ erigena, stratification decreased germination below that ed seed was 50-60 %. of the controls and had no significant effect on the In Gaultheria antipoda, G. crassa and G. depressa, germination of E. ciliaris. However, G. macrostigma, G. there was an initial period of two to three weeks with no macrostigma x depressa from Flagstaff, and the high germination, followed by a period of rapid germination altitude population of G. depressa showed a significant with germination .more or less completed in 30 days (G. increase in percentage germination and germination rate crassa), 40 days (G. antipoda) or 80 days (G. depressa). when stratified (Fig. 1). Different collections of G. The high altitude collection of G. depressa (1580m) began to germinate after only four weeks and showed an initial phase of rapid germination which was completed after about 90 days and secondary burst of germination Table 1. Percentage gennination of untreated seed which commenced at 120 days and was more or less sown in the cabinet and in the glasshouse, complete after about 290 days. Seed of G. macrostigma and of seed pretreated with 375 days of tended to show a more gradual increase of germination dark, moist storage in the cabinet and then over a longer period with germination of control seed sown in the glasshouse. tailing off after 200 days. The seed collected from the glass- dark/glass- Hawkdun Range showed poor germination (< 25 %) but Species cabinet house house that from Flagstaff showed > 60 % germination. The E. lusitanica 11.o· 67.o· 93.5' hybrid, P. macrostigma x depressa, showed some characteristics of its parents, and resembled G. depressa G. antipoda 78.0 63.0 77.0 in completing most of its germination between day 20 G. depressa (775m) 78.0 90.0 50.5' and day 100, although its final germination was low G depre~sa (1140m) 89.5 90.5 93.5 (< 30 %in control seed). Germination percentages in G. G. depressa (1580m) 78.5 82.0 82.5 macrostigmax crassa were similar to those of both its G. macrostigma (Hawkdun) 9.0' 45.0 29.5 parents, although germination rates were intermediate G. macrostigma X depressa 44.5 36.0' 59.0 between those of its parents. G. parvula showed very G. parvula 0.5 12.5 3.0 low germination (< 10% in untreated seed) with • denotes values are significantly different (P < 0.05)

Seed Symposium 1999 85 Germination aspects of some NZ Ericaceae macrostigma and G. macrostigma X depressa showed a were killed at 100°C (although some seeds of E. ciliaris variable response so that there was no overall significant germinated after treatment at 100°C for one minute). increase in germination percentage in stratified seed (Fig. When dry seeds of Gaultheria spp. and E. lusitanica 2), although there was a significant increase in were exposed to two minutes or 30 seconds at 100°C germination rate (Moore, 1996). there was generally no stimulation of germination. Five minutes of moist heat pretreatment of imbibed Germination percentages were depressed in low altitude seeds at 50°C (E. ciliaris) and 80°C (E. erigena) G. depressa (100°C for 2 minutes) and in G. parvula stimulated germination above that of controls and seeds (both treatments) (Fig. 3).

-control k , >+ j.stratified c 80 0 ~c 60 "§ Q) 40 Cl 20

El us Ecil Eerig Ga Gc Gd7 Gd11 Gd15 GmF GmH Gxd Gp Species Figure 1. Percentage germination of control seeds and seeds stratified for four weeks at 4°C. Species: Erica lusitanica (Eius), E. ciliaris (Ecil), E. erigena (Eerig), Gaultheria antipoda (Ga), G.crassa (Gc), G. depressa from 775 m (Gd7), 1140m (Gdll) and 1580m (GdlS), G. macrostigma from Flagstaff (GmF) and Hawkdun Range (GmH), G. macrostigma x depressa (Gxd), G. parvuw (Gp).

100 control

. stratified 80 ~ c 0 - m ""c 60 ·e.... Cl> C> 40

~0 20

0 F Dec 92 F Mar 93 F May 93 H Apl93 S Apl 93 x depressa Seed lots

Figure 2. Percentage germination of control seeds of Gaultheria macrostigma and G. macrostigma x depressa and of seeds stratified for four weeks at 4°C. Seeds collected on different dates from Flagstaff (F), Hawkdun Range (H), and Silverpeaks (S).

Seed Symposium 1999 86 Germination aspects of some NZ Ericaceae 100 -control 80 -h1 c: 0 c:::::J h2 ~ 60 .Ec: .... (J) 40 Cl

0~ 20

0 El us Ecil Eerig Ga Gc Gd7 Gd11 Gd15 GmF GmH Gxd Gp Species

Figure 3. Percentage gennination of control seeds and dry seeds subjected to 30 sec (h1) and two min (h2) at 100°C or imbibed seeds subjected to 5 min at either 60°C (h1, Ecil) or sooc (hl, Eerig) and one minute at 100°C (h2, Ecil,Eerig). Species: Erica lusitanica (Elus), E. cilwris (Ecil), E. erigena (Eerig), Gaultheria antipoda (Ga), G. crassa (Gc), G. depressa from 775 m (Gd7), 1140m (Gdll) and 1580m (Gdl5), G. macrostigma from Flagstaff (GmF) and Hawkdun Range (GmH), G. macrostigma x depressa (Gxd), G. parvula (Gp).

After-ripening, longevity, and seasonal variation germination of seeds stored for only one month was Dry seed of E. lusitanica and Gaultheria spp. was significantly less than that stored for longer periods. stored for 1, 3, and 16 months and germinated in the Seed collected in different years (Table 3) did not cabinet (Table 2). Germination of most species was always show the same percentage germination. unaffected by 16 months of dry storage; only seed of G. Germination was significantly greater in seed collected in crassa showed reduced germination. However all 1993 for G. antipoda, G. depressa (715 m), G. collections of G. depressa and G. depressa x macrostigma (Hawkdun) and G. macrostigma Xdepressa, macrostigma showed evidence of after-ripening as while seed of G. macrostigma from Flagstaff showed

Table 2. Percentage gennination of untreated seed Table 3. Percentage germination of untreated seed following dry storage of one, three and collected in autumn 1992 and autumn 1993 sixteen months in standard conditions in and genninated in standard conditions in the growth cabinet. the growth cabinet. 3 16 1992 1993 Species month months months E. lusitanica 14.5 7.0 E. lusitanica 14.5 10.0 17.0 G. antipoda 60.5 84.0' G. antipoda 60.5 64.0 78.0 G. crassa 85.5 77.5 G. crassa 85.5 96.5' 78.0 G. depressa (775m) 65.0 91.0' G. depressa (775m) 65.oa 92.5 89.5 G depressa (1140m) 76.5 87.0 G depressa (1140m) 76.5 94.5 89.5 G. depressa (1580m) 55.5 75.5 G. depressa (1580m) 55.5' 80.0 78.5 G. macrostigma (Flagstaff) 77S 43.0 G. macrostigma (Hawkdun) 15.5 24.0 9.0 G. macrostigma (Hawkdun) 15.5 62.5' G. macrostigma X depressa 9.5' 28.5 44.5 G. macrostigma X depressa 9.5 28.5a G. parvula 5.5 7.0 0.5 G. parvula 5.5 1.5 a denotes values are significantly different (P < 0.05) a denotes values are significantly different (P < 0.05)

Seed Symposium 1999 87 Germination aspects of some NZ Ericaceae Table 4. Percentage germination after 0, 1, 3 and was broken by low temperature stratification (Mallik and 10 cycles of hydration and dehydration. Gimingham, 1985). In this paper, germination in some seed lots of Gaultheria species (e.g., G. macrostigma, G. Number of dehydration cycles depressa and G. macrostigma X depressa) was stimulated by low temperature stratification whereas that of others 0 1 3 10 (including G. antipoda and G. crassa) was not. In G. E.lusitanica 7 2 4 8 macrostigma the responses were variable, apparently G. antipoda 84 72 82 84 depending on both location and season of collection (Fig G. crassa 78 78 76 66 2). Chilling requirements are often inferred from G. depressa - 1140m 87 88 82 80 observations of enhanced germination after over­ G. macrostigma (Flagstaff) 43 38 42 50 wintering (Fountain and Outred, 1991) but only G. macrostigma Xdepressa 29 28 14 32 experiment with adequate controls will establish a G. parvula 2 0 0 0 chilling requirement. Other factors such as prolonged moist storage, after-ripening, scarification of the seed coat by microbial or other agencies, and increased temperatures in spring could account for increased germination after overwintering. Burrows (1994a,b) significantly higher germination in the 1992 collection. suggests that few native New Zealand woody species G. macrostigma x crassa showed no difference between have a chilling requirement, reflecting their tropical years of collection (Moore, 1996). origins. Some of the native Gaultheria species, however, All species, except perhaps G. parvula which showed go against this trend and their variability of response no germination in any dehydration treatment but only reflects that found in ericaceous species from the 2 % germination in the untreated controls, were Northern Hemisphere. unaffected by up to ten repeated cycles of hydration and High temperature responses have generally been dehydration (Table 4). examined in species that inhabit fire-prone communities. Whittaker and Gimingham (1961) showed that short exposures to temperatures between 40 and 80°C Discussion stimulated germination in imbibed seeds of Calluna Stimulation of germination by light is a common vulgaris. Similar results were found for Erica cinerea characteristic of plants that form seed banks (Pons, 1991) (Bannister, 1965), E. ciliaris (Rose et al., 1996) and E. and is found in many ericaceous seeds such as Calluna erigena (Fig. 3). Mallik and Gimingham (1985) treated vulgaris (Gimingham, 1960; Pons, 1989a), Erica ciliaris dry seed and found that germination of Vaccinium (Rose et al., ·1996), E.cinerea (Bannister, 1965), E. myrtillus and V. vitis-idaea was stimulated by heat shock tetralix (Bannister, 1966; Pons, 1989), Pernettya (30 seconds at 100°C) but not by longer periods of mucronata (Arena et al., 1994), and Vaccinium myrtillus exposure. None of the dry seeds treated in this study (Grime et al., 1981) and the species examined in this (Fig. 3) showed any stimulation of germination after paper. Low temperatures may substitute for the light similar periods of heat shock and any significant requirement in some ericaceous seeds such as Arbutus diminution of germination was only slight. In the field, unedo (Ricardo and Veloso, 1987) and E. lusitanica germination of E. lusitanica is stimulated by fire (Mather (Mather and Williams, 1990). and Williams, 1990) but it is difficult to determine Ericaceous seeds do not necessarily show increased whether this is due to heat stimulation or other factors germination after low temperature stratification. Calluna such as removal of litter and disturbance exposing the vulgaris, Erica tetralix (Mallik and Gimingham, 1985), buried seeds to light. E. cinerea (Bannister, 1995), E. ciliaris (Rose et al., All seeds that were subjected to hydration and 1996) and E.lusitanica (Fig. 1) show no response to low dehydration were unaffected by up to ten such cycles. temperature stratification, whilst germination of E. Seed of Calluna vulgaris and Erica tetralix (Pons, 1989) erigena is depressed by it (Fig. 1). However, and of Leptospermum scoparium (Mohan et al., 1984) overwintered seed of E.lusitanica showed higher was similarly unaffected. Seeds often show enhanced germination than fresh seed (Mather and Williams, germination after a period of dry storage (after-ripening). 1990), germination in Pernettya (Gaultheria) mucronata Grime et al. (1981) found such enhancement in the was stimulated by 300 hours of chilling (Arena et al., majority of 403 species examined, particularly in small­ 1994) and dormancy in seed of Vaccinium vitis-idaea seeded species, but in this study only G. depressa and G.

Seed Symposium 1999 88 Germination aspects of some NZ Ericaceae macrostigma X depressa showed any evidence of after­ Bannister, P. 1965. Biological Flora of the British Isles: ripening. All species showed substantial germination Erica cinerea L. Journal of Ecology 53, 537-542 after 16 months of dry storage. and only G. crassa Bannister, P. 1966. Biological Flora of the British Isles: showed significantly reduced germination. Comparative Erica tetralixL. Journal of Ecology 54, 795-813. Bannister, P. 1990. Seed germination in Gaultheria data are hard to find. Dry seed of Erica lusitanica did antipoda, G. depressa, and Pernettya macrostigma. New not lose viability when stored for five months in the dark Zealand Journal of Botany 28, 357-358. (Mather and Williarns, 1990) and that of Leptospermum Bannister, P. and Jameson, P.E. 1994. Germination physi­ scoparium retained viability for one year (Mohan et al., ology of seeds from New Zealand native plants. In Seed 1984). Substantial germination occurred after two years' Symposium: Seed Development and Germination (eds. storage in Erica cinerea (Chippendale and Milton, 1934) Coolbear, C.A. Cornford and K.M. Pollock), pp 9-15, and after four and a half years in E. tetralix (Bannister, Agronomy Society of New Zealand, Special Publication 1966). Different collections of seed from the same 9. source may show differences in germination with respect Burrows, C.J. 1994a. Germination experiments with seeds to season and year. These doubtless reflect differences from the native New Zealand woody flora. In Seed Symposium: Seed Development and Germination in maturation and seasonal climate. (eds. P. Coolbear, C.A. Cornford and K.M. Pollock), pp There are many similarities between the germination 17-23, Agronomy Society of New Zealand, Special of Northern Hemisphere Ericaceae and the Southern Publication 9. Hemisphere species of Gaultheria examined here. All Burrows, C.J. 1994b. The seeds always know best. New germinate best in the light and germination in most Zealand Journal of Botany 32, 349-363. species is usually completely inhibited in the dark. Chippendale, H.G. and Milton, W.E.J. 1934. On the viable Viability is not.lost to any great degree in either dry or seeds present in the soil beneath pastures. Journal of moist storage, and imbibed seeds are unaffected by short Ecology 22, 508-531. periods of dehydration. These characteristics favour Fountain, D.W. and Outred, H.A.. 1991. Germination persistence in a seed bank until disturbance exposes the requirements of New Zealand native plants: a review. New Zealand Journal of Botany 29, 311-316. seed to light which in turn stimulates germination. Gimingham, C.H. 1960. Biological Flora of the British Although there is some evidence that low temperature Isles: Calluna vulgaris (L.) Hull. Journal of Ecology 48, stratification enhances germination in some seed 455-483. collections, substantial germination occurs in most Gimingham C.H. 1992. The Lowland Heathland species in its absence. While germination of dry seed is Management Handbook. English Nature, Peterborough. not stimulated by heat shock, there is little loss of Grime, J.P., Mason, G., Curtis, A.V., Rodman, J., Band, viability in dry seed subjected to two minutes at 100°C, S.R., Mowforth, M.A.G., Neal, A.M., Shaw, S. 1981. A suggesting tolerance to fire, which is not an uncommon comparative study of germinatien characteristics in a feature of scrubland. These seeds differ from those of local flora. Journal of Ecology 6!}, 1017-1059. other woody plants in the New Zealand flora - the Hill, M.O. and Stevens, P.A. 1981. The density of viable seeds in soils of forest plantations in upland Britain. selection of which has been described as "... affected Journal of Ecology 69, 597-613. neither by dark/light differences or by winter cold.... " Mallik, A.U. and Gimingham, C.H. 1985. Ecological (Burrows, 1994a). The similarities in germination effects of heather burning. II. Effects of seed between European and antipodean Ericaceae suggest that germination and vegetative regeneration. Journal of taxonomic affinities have overridden any such adaptation Ecology 73, 633-644. to local climates. Mather, L.J. and Williams, P.A. 1990. Phenology, seed ecology, and age structure of Spanish heath (Erica lusitanica) in Canterbury, New Zealand. New Zealand References Journal of Botany 28, 207-215. Arena, E.A., Pastur, M.P. and Vater, G. 1994. The role of Middleton, D.A. 1991. A critical examination of the status stratification and light on seed germination in Pernettya of Pernettya as a genus distinct from Gaultheria. mucronata (L. f.) Gaudlich ex. G. Don. New Zealand Edinburgh Journal of Botany 47, 291-301 Journal of Botany 32, 337-339. Mohan, E, Mitchell, N. and Lovell, P. 1984. Environmental Armstrong, J.M. 1964. Cytological studies in some New factors controlling seed germination of Leptospermum Zealand plants. Unpublished MSc Thesis, University of scoparium (manuka). New Zealand Journal of Botany Otago, Dunedin, New Zealand. 23, 95-101.

Seed Symposium 1999 89 Germination aspects of some NZ Ericaceae Moore, S.H. 1996. Germination characteristics of some Rose, R.H, Bannister, P. and Chapman, S.B. 1996. native and naturalized species of Ericaceae in New Biological Flora of the British Isles: Erica ciliaris L. Zealand. Unpublished MSc thesis, University of Otago, Journal of Ecology 84, 617-628. Dunedin, New Zealand. Whittaker, E. and Gimingham, C.H. 1962. The effects of Pons, T.L. 1989. Dormancy and germination of Calluna fire on the regeneration of Calluna vulgaris (L.) Hull vulgaris (L.) Hull and Erica tetralix L. seeds. Acta from seed. Journal of Ecology SO, 815-822. Oecologica/Oecologia Plantarum 10, 35-43. Pons, T.L. 1991. Induction of dark dormancy in seeds: its importance for the seed bank in the soil. Functional Ecology 5, 669-675. Ricardo, C.P.P. and Veloso, M.M. 1987. Features of seed germination in Arbutus unedo L.. In Plant Response to Stress (eds. J.H. Tenhunen, F.M. Catarino, O.L. Lange, and W.C. Oechel), pp 565-572, NATO ASI Series G, Ecological Sciences Vol. 15. Springer Verlag, Berlin.

Seed Symposium 1999 90 Germination aspects of some NZ Ericaceae